Line pattern retrieval using relational histograms

Huet, Benoît; Hancock, Edwin R.

doi:10.1109/34.817414

Cited by 61 publications

(51 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Future direction is to speed up the FPD computation further by using approximation methods (Huet and Hancock, 1999;Pavlou and Allinson, 2009) and pre-filtering the whole database using histogram-based indexing (Grauman and Darell, 2004;Berg et al, 2005). Another direction is to use the proposed system to get the best match for the crime scene mark and use that to interpret the result in terms of the strength of the evidence it can support to find if the suspect is the offender.…”

Section: Discussionmentioning

confidence: 99%

An efficient clustering-based retrieval framework for real crime scene footwear marks

2012

View full text Add to dashboard Cite

Abstract:As the most common type of evidence at crime scenes, footwear marks are found more often than fingerprints, and yet left largely unused due to lack of efficient and reliable tools. While the central task is stated simply -retrieve the closest matches among a database of known outsole prints -the difficulty is the poor quality of the marks and a very large and increasing number of outsole patterns. Since grouping the database into clusters can dramatically speed-up retrieval, we propose clustering based on recurring outsole patterns. The clustered database is used to retrieve similar prints for a given crime scene mark. Geometric shapes like line segments, circles and ellipses are proposed as features for crime scene marks. Then these features are structurally represented in the form of an attributed relational graph (ARG). Robust ARG matching is achieved with the introduced footwear print distance (FPD), a similarity measure for footwear prints. Sensitivity analysis of FPD is performed to show its robustness. The proposed system is invariant to scale, translation, rotation and insensitive to noise and degradations of the prints. Experiments show that the approach outperforms other state-of-the-art footwear print retrieval systems.

show abstract

Section: Discussionmentioning

confidence: 99%

An efficient clustering-based retrieval framework for real crime scene footwear marks

2012

View full text Add to dashboard Cite

show abstract

“…The choice of distance metric for measuring the similarity between histograms may affect recognition performance [8]. The L1 and L2 norms are both commonly used in histogram comparison.…”

Section: Attribute Histogramsmentioning

confidence: 99%

“…Histograms have proven to be simple and powerful attribute summaries which can be used to great effect in the recognition of objects from large image databases. The idea was originally popularized by Swain and Ballard, who used color histograms [20], and it has subsequently been successfully used for texture recognition [6], 3D object recognition from rangeimages using shape-index spectra [5], and for line-pattern recognition [8]. We explore several attribute representations derived from the needle-maps returned by SFS.…”

Section: Introductionmentioning

confidence: 99%

Object recognition using shape-from-shading

Worthington

Hancock

2001

IEEE Trans. Pattern Anal. Machine Intell.

Self Cite

View full text Add to dashboard Cite

AbstractÐThis paper investigates whether surface topography information extracted from intensity images using a recently reported shape-from-shading (SFS) algorithm can be used for the purposes of 3D object recognition. We consider how curvature and shape-index information delivered by this algorithm can be used to recognize objects based on their surface topography. We explore two contrasting object recognition strategies. The first of these is based on a lowlevel attribute summary and uses histograms of curvature and orientation measurements. The second approach is based on the structural arrangement of constant shape-index maximal patches and their associated region attributes. We show that region curvedness and a string ordering of the regions according to size provides recognition accuracy of about 96 percent. By polling various recognition schemes, including a graph matching method, we show that a recognition rate of 98-99 percent is achievable.

show abstract

“…Here we have used the method of Huet and Hancock [6] to compute pairwise attributes from the relative angles and lengths of the line-segments defining the characters.…”

Section: Line Patternsmentioning

confidence: 99%

Spectral Analysis of Complex Laplacian Matrices

Wilson

Hancock

2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper explores how to extend the spectral analysis of graphs to the case where the nodes and edges are attributed. To do this we introduce a complex Hermitian variant of the Laplacian matrix. Our spectral representation is based on the eigendecomposition of the resulting Hermitian property matrix. The eigenvalues of the matrix are real while the eigenvectors are complex. We show how to use symmetric polynomials to construct permutation invariants from the elements of the resulting complex spectral matrix. We construct pattern vectors from the resulting invariants, and use them to embed the graphs in a low dimensional pattern space using a number of well-known techniques including principal components analysis, linear discriminant analysis and multidimensional scaling.

show abstract

Line pattern retrieval using relational histograms

Cited by 61 publications

References 20 publications

An efficient clustering-based retrieval framework for real crime scene footwear marks

An efficient clustering-based retrieval framework for real crime scene footwear marks

Object recognition using shape-from-shading

Spectral Analysis of Complex Laplacian Matrices

Contact Info

Product

Resources

About